SUITMA 2000 Essen - Suitability of urban demolition soils in Sheffield for wild flower meadows
Topics > Events > SUITMA > SUITMA 2000 Essen > Papers and abstracts
| This paper is one of the Green areas, park and urban forest theme of the SUITMA 2000 Essen symposium. |
| Francesca Bretzel,i | James Hitchmough.ii |
- i - Department of soil chemistry, CNR Institute for ecosystem, Pisa, Italy
- ii - Department of landscape, University of Sheffield, Sheffield, United Kingdom.
Contents
Abstract
Soils associated with urban re-development sites generally consist of an infertile mix (subsoil, building material) which is gradually colonised by weedy native and exotic grasses and forbs adapted to these soil conditions. Research was carried out in Sheffield on a post industrial site to see whether a visually more attractive meadow-like vegetation of native and exotic species could be established by oversowing with only limited changes to the soil. The soil was analysed for chemical (pH, organic matter, NPK total and available) and physical aspects (bulk density, granulometry). The site was sown with 17 species of forbs with 6 different soil cultivation treatment. Analysis showed that the soil was alkaline (8.5) highly compacted (bulk density 1.6%) with low plant available nitrogen. Forb establishment was superior following a spring as opposed to autumn sowing. The most successful soil cultivation treatment was the combination of burning and cultivation
Introduction
The most biodiverse herbaceous vegetation is often found on infertile soils[1]. Under infertile conditions, the biomass of highly productive species is restricted, enabling slower growing species that would otherwise suffer competitive elimination to co-exist. In the past 30 years many soils have been improved in the countryside by fertiliser addiction and other forms of intensification leading to the loss of diverse vegetation. In contrast, in towns and cities soils made infertile by construction work (compaction and removal of topsoil) and industrial activity (deposition of metals and other contaminants) are common, and are seen as a problem because traditional horticultural vegetation such as turf and flowers grow poorly on these materials. These soils are however really a potentially valuable resource for the creation of species rich native vegetation in towns, to compensate for the loss of this vegetation in the countryside. The plant communities typically found on these types of soils have been reviewed by Gilbert[2] [3]. The aims of the study described in this paper was to see whether it was possible to establish a diverse, colourful meadow-like vegetation on highly infertile urban soils by over-sowing with a seed mix of native and exotic herbaceous perennials (forbs). Exotic (non-native) species were added in order to contrast their ecological needs with those of native species, and also to add addition colour and interest to the urban site in question. Current landscape practice on such sites in Britain is to spread a 300 mm layer of topsoil (transported from elsewhere) to provide a highly root penetrable and nutrient rich substrate which is then planted with shrubs and trees. Given the high densities of weedy species present in these topsoils as rhizome fragments and a soil seed bank, plus extremely limited post planting maintenance, on many of these sites the shrubs are often replaced after a few years by a tall weed community. This weedy vegetation is less attractive and generally less biodiverse than the vegetation that had spontaneously colonised the original infertile soils. For this reason the practice of topsoiling urban clearance sites is often undesirable, and suggests that using the existing on-site-substrates may be a far more sustainable approach where highly vigorous vegetation is not a pre-requisite.
Locations and Material
The research was carried out for two years on a former industrial site in Sheffield. The substrate consisted of a heterogeneous mix of compacted deep subsoil and crushed masonry. It supported a rather open community of ruderal weeds, dominated by legumes and grasses. The site was in the early stage of transition to taller herbaceous vegetation and ultimately pioneer woodland. The aim was to study the germination and establishment of a seed mix of 17 native and exotic forbs (herbaceous perennials), sown in autumn and spring, in relation to soil quality and a number of pre-sowing site treatments. Details of the species sown are given in Table 1. The site was divided into 48 x 1 m² (100 x 100 cm) treatment plots with 6 different preparation-soil cultivation treatments. The experimental design adopted was a fully randomised block, with 4 replicates of each treatment. The treatments were designed to involve only limited manipulation of the site, and were as follows: cut off the existing vegetation to soil level; cut off and lightly cultivate the surface (to 10 cm); spray vegetation with glyphosate herbicide (4L/ha CP in medium volume); spray glyphosate then cultivate; burn existing vegetation (using a propane fuelled heat gun); burn and cultivate. Half the blocks were sown in spring and the other half in autumn. Twenty five seeds of each species was counted out by hand on sown onto each plot, giving a sowing density of 425 seeds/m².
| Natural distribution : | Species : | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| North America | Asclepias tuberosa | Aster azureus | Aster laevis | Coreopsis lanceolata | Echinacea pallida | Liatris aspera | Ratibida pinnata | Solidago rigida | |
| Europe | Bupthalmum salicifolium | Centaurea scabiosa | Dianthus cartusianorum | Galium verum | Knautia arvensis | Malva moschata | Origanum vulgare | Primula veris | Salvia x superba |
The success of the various treatment combinations were assessed by identifying and counting the seedlings of species sown on each plot, in early summer (once seedlings were large enough to be conclusively identified) and in autumn. An assessment of weed cover plus the dominant weed species present was also made.
For soil chemical characteristics total and available NPK, Organic Matter and pH were analysed[4]. Samples were taken in the 0-20 cm depth layer of soil. For physical characteristics bulk density and granulometry were analysed[5] ; for bulk density the undisturbed samples were taken in the 5-30 cm depth layer, where the roots of forbs were well developed.
Results
The soil analysis results showed a soil with alkaline pH (8.5) typically due in urban sites to the high presence of calcium rich building materials. The organic matter content of the soil was surprisingly high as indicated by the values for total nitrogen, although plant available nitrogen was low. The latter is likely to be caused by the absence of microbes and the lack of aeration leading to denitification instead then mineralization. Phosphorous content was very variable, especially the plant available fraction (0.00-14.72 mg/kg), but while the total content corresponds to the average, the plant available was low. Values for potassium levels were fairly average and uniform across the different samples content.
| Total N
(mg/kg) |
Total P
(mg/kg) |
Total K
(mg/kg) |
Plant available NH4-N
(mg/kg) |
Plant available PO4-P
(mg/kg) |
Plant available K
(mg/kg) |
Organic matter
(%) |
pH |
|---|---|---|---|---|---|---|---|
| 1228 | 802 | 28821 | 0.693 | 10.14 | 165 | 6.8 | 8.01 |
Physically, the soil is highly compacted (bulk density average=1.66, range 1.54-1.79), even though the high percentage of crushed building rubble fraction (> 2mm) presumably improves drainage and oxygenation and facilitate root growth. The granulometric fraction reveals a high presence of sand, part of its origin is probably due to the finer crushed masonry; silt is in high percentage too, this is often related to difficulties with drainage and aeration.
| Silt (%) | Sand (%) | Clay (%) | Humidity (%) |
|---|---|---|---|
| 21.8 | 68.0 | 9.1 | 0.92 |
Taking species as a single factor, the success of field establishment of seedlings was shown to depend heavily upon pre-sowing treatment. The most successful treatments, in terms of mean numbers of seedlings for spring and autumn sowing (in decreasing order of success) were: burning followed by cultivation, herbicide followed by cultivation, and burning with no cultivation. Establishment of seedlings was most successful from a spring as opposed to autumn sowing. Species that established did however grow very slowly on this hostile soil, and some seedlings were lost due to competition with herbivore and spontaneously colonising grasses and forbs.
File:Summary of number of seedlings of sown species.png
Figure 1 Summary of number of seedlings of sown species in response to pre-sowing treatments. Values for 1999 represent surviving second year seedlings.
Discussion
As can be seen from the physical and chemical analyses the urban soil on this site presented considerable hostility in terms of compaction and low levels of plant available nitrogen. Although never quantified, observation across the two years of the experiment showed the site was extremely wet and presumably anaerobic in winter. By early summer the soil had however become a dry impenetrable material. It would not be recognised as a soil for growing plants from the perspective of agriculturists and horticulturalists. Despite this the experiment showed that it was possible to establish a diverse range of native and exotic forb species on this site given some initial control of the species naturally colonising the site and some disturbance of the surface to improve seed-soil contact. Given the size of the plots it was necessary to use a surrogate form of cultivation that would mimic the use of a chisel plough or similar implement on a larger site. Due to the presence of large pieces of building rubble in the profile of these soils, only shallow, essentially surface cultivation is possible. This was achieved on the experimental blocks by using a garden fork to fracture, but not invert the upper 100 mm of the profile. It is interesting to note however that burning alone also gave good results, probably as a result of destroying organic detritus, mosses and lichens which are known to be detrimental to the germination of many species. Where cultivation can not be employed it is possible that burning may provide an acceptable substitute.
Whilst satisfactory numbers of forbs were established, the growth of many species on this urban soil was extremely slow. Low nutrient levels, summer drought and winter and spring anaerobism are likely to have been important contributors to this, however competition with spontaneously colonising grasses and forbs, plus rabbit and mollusc action was probably equally or more important. The species that grew best were all European species associated with grassland habitats dominated by C3, winter growing grasses. These were Centaurea scabiosa, Knautia arvensis, Malva moschata, and Primula veris. A percentage of these species were large enough to flower in the second growing season. Species adapted to more continental, “gappy” dry grassland or winter dormant grasslands of C4 grasses (as in the case of the north American spp.) grew less well, presumably due to intensive competition in late winter and spring with the weedy colonists. Some species (Asclepias tuberosa, Liatris aspera, and Salvia x superba) proved to be very attractive to slugs and snails and by year 2 had largely been eliminated.
The experiment described in this paper has since been destroyed by the construction of a new industrial building. Security of site tenure is a recurrent problem in researching the development of slow growing herbaceous vegetation on destroyed urban soils. By their very nature these sites are often in transition between old and new uses. The success of this work however suggests that further research is warranted on using these types of urban soils as substrates for biologically diverse herbaceous vegetation.
References
Directly cited in the text
- ↑ Grime, J.P. (1979): Plant Strategies and Vegetation Processes. John Wiley, Chichester
- ↑ Gilbert O.L (1989): The Ecology of Urban Habitats. Chapman and Hall, London.
- ↑ Gilbert O.L. (1992):The Flowering of the Cities; the natural flora of urban commons. English Nature, Peterborough.
- ↑ Allen S.E., (1989): Chemical Analysis of Ecological Materials, 2nd edition, Blackwell Scientific Publications.
- ↑ ASA-SSSA (1996): Methods of soil analysis, Part 1 and 3-Physical and Chemical Methods-2nd edition ASA-SSSA, Madison, Wisconsin, USA.
Other references
- Bretzel, F. Petruzzelli, G.Pini, R.(1998): Chemical and physical problems in Urban Soils. VII International Congress of Ecology, INTECOL, Florence, p.58.
- Bretzel, F. Lubrano, L. Petruzzelli, G. (1998): Heavy metal contamination in urban soils, VI International FZK/TNO Conference on Contaminated Soils, p. 975-976.
- Bullock P, Gregory P.J (1991) Soils in the urban environment. Blackwell Scientific Publications.
